December 17, 1999

Breast cancer risk marker identified

Featured Image

A study by William Dupont, Ph.D., has found evidence of a biologic marker of increased breast cancer risk. (photo by Dana Johnson)

Breast cancer risk marker identified

Women with a benign breast disease whose cells lose the ability "to hear" signals instructing them not to divide are at increased risk of later developing breast cancer, Vanderbilt-Ingram Cancer Center researchers have found.

In the Dec. 15 issue of the Journal of the National Cancer Institute, the scientists report that a reduction in expression of a receptor for transforming growth factor beta (TGFb) occurs early in the progression from hyperplasia to malignancy in some women. Hyperplasia is a benign condition that involves an overgrowth of apparently normal cells.

"This is the first time we've found credible evidence of a biologic marker of increased breast cancer risk in women with unequivocally benign breast disease," said William Dupont, Ph.D., professor of Preventive Medicine.

A laboratory test to determine levels of expression of this receptor might be used in the future to reassure women at no increased risk or to guide decisions about screening and preventive measures in women whose risk is elevated, Dupont said. In addition, the findings suggest a new angle for scientists to explore in developing prevention strategies.

TGFb is a powerful regulator of cell growth whose job is to instruct cells not to divide. The cells "hear" this signal through the TGFb type II receptor, into which TGFb locks itself before sending its instructions to the cell nucleus.

This research builds on more than two decades of work by Dupont, Dr. David Page, and their colleagues with a group of nearly 10,000 women who underwent biopsies revealing benign breast disease at three Nashville hospitals over a 30-year-period.

Through their ongoing analysis and follow-up of the Nashville Breast Study Cohort — the largest of its kind — this research team has contributed greatly to the understanding and definition of benign breast disease.

"We have found that there are different types of benign breast disease with different levels of risk," Dupont said.

"Most women who have undergone bening breast biopsy — about 70 percent — have non-proliferative disease that's not associated with any elevation in risk. A minority, about 4 percent, have a four- to five-fold increase in risk, while the remainder have only a mild elevation of risk."

Thus far, indicators of breast cancer risk identified through this work have been based on histology — how the cells look under the microscope in terms of shape, size, distribution and so forth.

A few so-called molecular markers — abnormal expression of particular cancer-related genes like p53 or HER-2/neu, for example — have been identified that help predict prognosis in patients with well-established breast disease.

This latest work focuses on differences in risk among women with the same diagnosis of benign breast disease based on histology.

"Finding molecular markers of risk in women who are still very far away from developing breast cancer has been difficult," Dupont said.

"Breast cancer is a disease with a very long memory. Things that happen during a woman's second decade of life — age of first menstruation, age of first pregnancy — can affect a woman's risk of developing breast cancer decades later. There are things happening in the cell over a long period of time and it would nice to know what they are."

The scientists identified 54 women in this cohort whose biopsies, conducted between 1965-1978, revealed a diagnosis of epithelial hyperplasia lacking atypia and who subsequently developsed invasive breast cancer.

Each was matched to two "control" women with the same diagnosis, of similar age and whose diagnostic biopsy was done at about the same time. However, the women in the control group did not develop breast cancer during subsequent follow-up.

The researchers tested tissue removed at the time of each woman's initial biopsy with antibodies to the TGFb type II receptor. This test produces a brown stain in cells in which this receptor is active (expressed).

By analyzing expression of the TGFb receptor and subsequent development of invasive breast cancer, the researchers discovered that the fewer the cells able "to hear" TGFb's signals, the greater the cancer risk.

Specifically, when the receptor was expressed in fewer than 25 percent of the cells, the woman had more than three times the risk of subsequent invasive breast cancer, compared to women who had receptor expression in greater than 75 percent of their cells.

The findings "make sense" given what scientists know about TGFb's function, Dupont said.

"This also is exciting because ordinary hyperplasia is a common condition," he said. "This will help to subdivide between those who have no reason for increased concern and those who may have a moderate elevation in risk," he said.

"For these women, having regular mammograms might go from being a good idea to being a very good idea, and it might lower the age at which they might want to start. It adds another piece of information for women, their physicians and genetic risk counselors to take into account."

The work was supported by the National Cancer Institute and the Vanderbilt-Ingram Cancer Center. Dupont's colleagues include Dr. Helenice Gobbi, pathology fellow and first author on the JNCI paper; Dr. Jean Simpson, associate professor of Pathology and director of Anatomical Pathology; systems analyst W. Dale Plummer Jr.; research nurse Peggy Schuyler; Sandra Olsen, research instructor in Pathology; Dr. Carlos Arteaga, Ingram Professor of Cancer Research and director of the Vanderbilt-Ingram Cancer Center's Breast Cancer Research Program; and Page, professor of Pathology and Preventive Medicine.